I spent most of my in-library career just praying that I'd never have to catalog serials!
There are a number of different options/directions. If you are mainly interested in articles, then there are some fairly tried-and-true data elements that make up an article citation. In a way, you treat each article like a separate publication that just happens to need to be located within a journal with an issue, number and starting page number. The library approach to serials is more complex and may not be of interest to you. In library cataloging, the articles are not included, only the serial publications themselves. What gets complex about this is keeping track of the history of the serial: how it has changed name or publisher over many years; changes in ISBN (which is supposed to change when the title changes); changes in the name of the organization that publishes the journal; etc. What is tricky for OL I think is that scanning and digitizing takes place usually on bound journal volumes, and the article information isn't there. Then you have the issue of trying to link up an article (which is generally what people are looking for, not a whole volume) with the particular digitized volume. It gets worse when the physical, bound volume doesn't correspond to the "numbered" volume (like for a really thick journal where it's too big for an entire numbered volume to fit into a bound volume that then gets digitized). I think there are ways to simplify the problem and you definitely do NOT want to try to do it the way libraries do, which is overly complex. The experience that people have with the OpenURL (which is a way of linking articles to the journals themselves) can probably come in handy. All that said, a place to start would be: what is going to be the source of the metadata? There are huge databases of journal article citations. If you want to start bringing in that data, then you could begin by analyzing how those records might link to journals that the Archive has scanned. kc Quoting George Oates <[email protected]>: > Hi all, > > There's a slim chance I asked this list about serials cataloging last time I > poked at it, but, I've started looking at it again and wondered if > you could help. > > The time has come to seriously consider adding both multi-volume works and > serials cataloging to Open Library, and knowing that it's a > notorious cataloging > issue, I wanted to reach out to learn from your experience and > insight on this > issue. > > I've been doing a survey of somewhat random journals (ones I like + > science-y + > literary + bookshop visual scans) to see if I can isolate any > consistent fields. > Along with the recent "Minimum Viable Record" post I did on the OL > blog [1], I'm > searching for a minimum set of fields we could enlist to describe serials > somewhat generically. Seems like there are: Title, Date, Volume, > Issue, but non > of these are used consistently, as I'm sure you're aware. > > I've looked at the LoC Serials Cataloging Issues site [2], which is uber > technical. I did however find a useful "Catalogers Cheat Sheet" > (PDF) which was > a good overview of MARC handling stuff [3]. > > So, I'm wondering if any of you happen to be serials cataloger, or know of > anyone who might be interested to talk with Open Library about how > we might do > this well... Or, if you know any useful web-based resources, I'd love links! > > Cheers, > george > > [1] http://blog.openlibrary.org/2011/04/11/minimum-viable-record/ > [2] http://www.loc.gov/acq/conser/issues.html > [3] PDF: http://www.loc.gov/acq/conser/pdf/CheatSheetforCSR.pdf > _______________________________________________ > Ol-discuss mailing list > [email protected] > http://mail.archive.org/cgi-bin/mailman/listinfo/ol-discuss > To unsubscribe from this mailing list, send email to > [email protected] > -- Karen Coyle [email protected] http://kcoyle.net ph: 1-510-540-7596 m: 1-510-435-8234 skype: kcoylenet _______________________________________________ Ol-discuss mailing list [email protected] http://mail.archive.org/cgi-bin/mailman/listinfo/ol-discuss To unsubscribe from this mailing list, send email to [email protected]
